ML062050012

Comanche Peak, Attachment 1 to TXX-0612 Response to Requests for Additional Information for the Replacement Steam Generator Supplement to Txu Power'S Large and Small Break Loss of Coolant Accident Analysis Methodologies, Pages 1 Through 74

ML062050012
Person / Time
Site:	Comanche Peak
Issue date:	07/17/2006
From:	TXU Power
To:	Office of Nuclear Reactor Regulation
References
CPSES-200601350, TAC MC6899, TXX-06125
Download: ML062050012 (74)

Text

Attachment I to TXX-06125 Page 1 of 216 ATTACHMENT I TO TXX-06125 RESPONSES TO REQUESTS FOR ADDITIONAL INFORMATION FOR THE REPLACEMENT STEAM GENERATOR SUPPLEMENT TO TXU POWER'S LARGE AND SMALL BREAK LOSS OF COOLANT ACCIDENT ANALYSIS METHODOLOGIES July, 2006 Attachment I to TXX-06 125 Page 2 of 216 TABLE OF CONTENTS I. INTRODUCTION ii. LOCA QUESTIONS AND RESPONSES III. NRC RAI REFERENCES IV. COMPLETE SETS OF PLOTS FOR ADDITIONAL CALCULATIONS V. MISCELLANEOUS FIGURES Attachment 1 to TXX-06125 Page 3 of 216 I. INTRODUCTION This document provides responses to the NRC request for additional information (RAI) in connection with the topical report (Ref. 3): ERX-04-004, "Replacement Steam Generator Supplement to TXU Powver's Large and Small Break Loss of Coolant Accident Analysis Methodologies." The format of the responses is as follows: In the section entitled "LOCA QUESTIONS AND RESPONSES," each NRC LOCA-related question is reprinted and followed by its response.

Any references within the NRC questions themselves are listed at the end of that section. That list is the same list of references the NRC submitted with the RAIs. Whenever new references are needed for a response, these are provided within the space for the response, just prior to it. These additional references are labeled according to the question number where the reference first appears. For example, Reference Q.7.g.(l) appears at the top of the response to Question 7, item (g), sub-item (1).The section "COMPLETE SETS OF PLOTS FOR ADDITIONAL CALCULATIONS," contains complete sets of plots for all the additional runs performed to address the LOCA RAI questions.

Each complete set is comprised of the original set of variables submitted inl the topical (Ref. 3) as well as the set of variables wvhose plots wvere requested in Question 7.a (Q.7.a). A total of 180 (10 sets of 18) plots are provided.The content of the remaining sections, "MISCELLANEOUS FIGURES" is self-explanatory, presenting figures which are not part of the complete sets of plots for the various runs.Finally, as described at the top of page 2-1 of Reference

[10] and in the NRC SER attached to it, the CPSES Small Break LOCA methodology is essentially an application of the SPC methodology of Ref.Q.7.g.(l), (also Ref. 1.1 of Reference

[10]). None of the models have been changed with respect to the original SPC methodology.

Attachment I to TXX-06 125 Page 4 of 216 IH. LOCA QUESTIONS AND RESPONSES The first LOCA question on the RAI list is Question number 7.7. With respect to the changes in CPSES's small break loss of coolant accident (SBLOCA) methodology described in ERX-04-004, "Replacement Steam Generator Supplement to TXU Power's Large and Small Break Loss of Coolant Accident Analysis Methodologies," (Ref. 3), provide the following information.

a. As part of its break spectrum sensitivity study, the licensee considered break sizes of 3, 4, and 5 inches. Provide the following plots for the 3, 4, and 5 inch breaks (and all other requested break sizes) as part of the SBLOCA analysis submittal.

I. Two-phase mixture level in thme core vs. time, HI. Downcomer liquid level vs. time, Ill. Steam temperature at the peak clad temperature (PCT) location vs. time, IV. Heat transfer coefficient vs. time at the PCT location, V. Condensation rate in the cold leg discharge injection sections vs. time, VI. Break quality vs. time.TXU Power Response: The additional plots for the 3", 4" and 5" breaks are provided, along with the original sets submitted with Reference

[3], in the section: "COMPLETE SETS OF PLOTS FOR ADDITIONAL CALCULATIONS".

Those are sets: A-I through A-I18 (3" break), B-i through B-18 (4" break), and C-A through C-18 (5" break). That section also provides the plot sets of Reference

[3] as wvell as the additional plots requested in Q.7.a, for all additional cases run in connection with this RAI.b. Please provide the limiting top peaked axial power distribution for each break size analyzed.TXU Power Response: The limiting top peaked axial power distribution for each cycle is obtained as described on page 3-3 of Reference

[10]. As explained on page 4-3 of Reference

[3], the limiting axial power distribution for the Reference

[3] calculations was obtained applying that approach to CPSES-1 cycle 11.A plot of the actual limiting power shape for CPSES 1 cycle 11, "NOM Shape # 948", is Figure 1, provided in the section: "MISCELLANEOUS FIGURES."

Attachment I to TXX-06 125 Page 5 of 216 c. Analysis of integer break sizes does not assure the limiting small break has been identified.

The SBLOCA limiting break is typically a break that depressurizes to an RCS pressure just above the accumulator actuation pressure, so that only high pressure safety injection (HPSI) terminates the PCT. The PCT for the limiting 4-inch break size was terminated by accumulator injection.

As such, please provide an analysis of that break which depressurizes to an RCS pressure just slightly above the accumulator actuation pressure and show that this is not the limiting SBLOCA.Break sizes which differ by as little as 0.005 ft 2 can result in increases in PCT in excess of 50 F for break sizes in the ranges 2 -5 inches. As necessary, please perform analysis of breaks between 3 and 4 inches, and 4 and 5 inches to assure the limiting break has been identified.

Provide the analysis and a full complement of plots for each break size, including the limiting break.TXU Power Response: In wvhat follows, it should be kept in mind that CPSES has a relatively high accumulator set point pressure (around 600 psia).A CPSES SBLOCA that depressurizes to an RCS pressure just above accumulator set point pressure wvould have to be smaller than 3". This is because the 3" break presented in Reference

[3] was also terminated by accumulator injection, i.e. it was accumulator injection that led to the clad temperature turn around for the 3", the 4" and the 5" inch breaks. Yet, the 3" break PCT is lower than the 4", which was the limiting break for the analyses of Reference[3]. The logical conclusion from these observations is that the CPSES SBLOCA limiting analysis is terminated by accumulator injection.

This is re-assuring because such an outcome is conservative relative to the scenario postulated in Q.7.c. This is because the transient would have terminated sooner (i.e. at a higher RCS pressure) and consequently have had a lower PCT, had it terminated due to pumped EGGS rather than accumulator injection.

In order to provide further assurance that the limiting CPSES SBLOCA analysis is (conservatively as explained above) terminated by accumulator injection, consider the analysis of a 1.5" break conducted to address Q.7.d below. That analysis showed that a 1.5" break was still not small enough to prevent accumulator discharge (set F-I through F-I18).Therefore, a break for which the accumulators do not inject would have to be smaller than 1.5" and as that analysis shows, such breaks are not limiting.Between 3" (0.049 ft 2 ) and 4" (0.087 ft 2 ) there are 7 break sizes differing by 0.005 ft 2 in area and between 4" and 5" (0.136 ft 2 ) there are 9 break sizes differing by 0.005 ft 2 in area.Therefore covering this entire range with such level of detail would require the analysis of 16 break sizes between 3 and 5 inches. This seems like an unreasonable burden given the margin between the large break and the small break LOCA PCT and the demonstration that the CPSES accumulators, given their high set point pressure, will discharge for breaks larger than 1.5.During the March 29 teleconference the NRC clarified that the request for additional fractional break sizes would be met by TXU providing a couple of additional fractional break analysis results in the 3" to 5" range that would bracket the limiting break. TXU is presenting Attachment 1 to TXX-06 125 Page 6 of 216 two such breaks, since they confinm that the 4" break is limiting.

One break is between 4" and 5" inches and another is betwveen 3" and 4". The size for these fractional size bracketing breaks was determined by starting with the 4" limiting integer break and increasing and decreasing the break size by 2*0.005 ft 2 = 0 .01 ft 2 in area. This process resulted in break diameters of 4.22" (0.097 ft 2 ) and 3.76" (0.077 ft'). The PCTs for these breaks were 1555OF for the 4.22" and 1530OF for the 3.76". A full complement of plots for these breaks is provided in the section: "COMPLETE SETS OF PLOTS FOR ADDITIONAL CALCULATIONS".

Set D-l through D-18 is for the larger (4.22") break and set E-l through E- 18 is for the smaller (3.76") break.Since TXU performs large and small break LOCA analyses for every fuel cycle reload, TXU could analyze fractional break sizes around the limiting integer size break for any cycle for which the small break LOCA PCT falls wvithin 25 0 F of the large break LOCA PCT (or if it is greater by any amount). This would be analogous to an existing provision in the methodology that triggers a cycle-specif ic time step study (Section 3.2.3 of Reference

[10]).During a June 14 phone call the NRC asked TXU to confirm that the oxidation acceptance criteria of 10 CFR5O.46 for SBLOCA are addressed in the methodology.

Also, a concern was brought tip that is allayed by what has been discussed above about the CPSES SBLOCA terminating by accumulator injection.

The concern was the theoretical possibility of the core remaining uncovered for a long time wvith clad temperatures below the PCT, but high enough to challenge the oxidation acceptance criteria.

These requests are addressed in the following:

As showvn in Figures A-7, B-7, C-7, D-7, E-7, G-7, 11-7, 1-7, J-7, for breaks of 3", 3.76", 4"1, 4.22", 5" and various sensitivities, the CPSES SBLOCA clad temperatures increase nearly monotonically' and rapidly after CHF in all cases. Mechanisms that might keep the rods at relatively high but constant temperatures, i.e., that would prevent rod temperatures from increasing rapidly in post CHF heat transfer mode are not operative during CPSES SBLOCA for two main reasons: (1) System depressurization rates are relatively fast, even for the smallest breaks, e.g.the 3" break shown in Figure A-1. Even for the smallest breaks, the ultimate rod quench is due to accumulator injection.

Thus, the clad temperature turns around very qulickly once the relatively high accumulator set point pressure is reached (again see Figure A-I for the 3" break).(2) The limiting powver shapes peak very near the top of the core. For the temperatures to linger at relatively high values, the uncovered portion of the core would have to be above the peak node, but below the top of the core. When the level is below the peak power node, the temperature increases at the rates seen in the smaller breaks, e.g., the 3" break shown in Figure A-7. Obviously, if the level is above the top of the core Inflections in the increase can occur due to water redistribution in the core, as discussed in Q.7.1 belowv, but this is not long lived Attachment 1 to TXX-06125 Page 7 of 216 there wvill be no clad temperature increase.

But also, due to the power shape, thle region of potential concern is very small (12-10.25

= 1.75 ft), contains almost no power and therefore even if the level lingered there, there wvould be no heat up.If CPSES SBLOCA clad temperatures wvere susceptible to remaining for relatively long periods of time at relatively high temperatures, this wvould be seen in the smaller breaks.However, the smallest breaks do not showv this behavior at all. (See Figures A-7 and E-7 for the 3" and 3.76" breaks, respectively).

Thus, once in CHF, rod temperatures take off and do not turn around until quenched'.

As a result of this characteristic, the peak SBLOCA local oxidation has always coincided with the peak clad temperature (PCT). The table below illustrates this point for the relevant cases presented in this submittal.

Break Size PCT (OF) OXIDATION'

(%)3.00 inch 1226 0.07 (0.008)3.76 inch 1530 0.37 (0.036)4.00 inch 1830 1.75 (0.211)4.22 inch 1555 0.38 (0.03 8)5.00 inch 1236 0.05 (0.004)Finally, the SBLOCA methodology of Reference

[3] does address thle oxidation acceptance criteria of 10 CFR 50.46 (b) (2) and 10 CFR 50.46 (b) (3), as shown for example on page 4-2 of Reference

[10] and in the table above.2 Local Transient Oxidation.

This value is added to the initial pre-transient steady-state oxidation, typically 8%, for comparison to the 17% limit specified in 10 CFR 50.46 (b) (2). Values in parenthesis are the total hot pin oxidation, which represent an tipper bound to the values wvhich must meet the 1 % limit specified in 10 CFR 50.46 (b) (3).

Attachment 1 to TXX-06 125 Page 8 of`216 d. Please show the HPSI head vs. flow injection capability for the case for a severed emergency core cooling system (ECCS) injection line. Also show the analysis and a full complement of plots for this break size.TXU Power Response: The HPSI head vs. flow for the case of a severed EGGS injection line transmitted to TXU by Westinghouse in Reference 2.12 of Reference

[10] is the basis for the EGGS flowv versus head used in the GPSES SBLOGA analyses.

Per Reference 2.12 of Reference

[10], this flowv constitutes

"... the minimum total flow through all SI branch lines, excluding the highest flow line. The highest flow line is assumed to have ruptured and wvill sp~ill its flow". The EGGS flow vs. head for each loop used in the GPSES analyses is given in Table 2.6 of Reference[ 10].As described in Section 2.4.1.3 of Reference

[10], the GPSES EGGS is comprised of: (1) the centrifugal charging puImps/safety injection system (GGP), (2) the high head safety injection system (HHSI), (3) the low head residual heat removal system (RHR) and (4) the accumulators.

The HHSI line injects into the accumulator line which in turn connects to the cold leg. The accumulator line is a 10" line. The HHSI line is an 8" line for loops 2 and 3 and a 6" line for loops 1 and 4. The 10" and 8" sizes were not analyzed because they are beyond the SBLOGA scope, i.e., such breaks have large break rather than small break characteristics, e.g. none or very short-lived loop seal plugging, and therefore are not expected to be limiting, either in comparison with smaller breaks where the pressure hangs higher thereby minimizing EGGS injection or with respect to the large break LOGA, which maximizes inventory loss.Regarding the 6" break, Reference

[3] shows that the 5" break is already non-limiting due to the fact that the pressure drops very quickly to the accumulator set point for that break size.The RCS pressure would reach that set point even sooner for the 6" break. Note that GPSES accumulator set point pressure is relatively high (around 600 psi). Note also that, if the HHSI line were severed this would only cause loss of part of the broken loop EGGS, namely the HHSI flow. The remaining EGGS flow coming from the GCPs, enters the accumulator line directly at another location and wvould therefore not spill directly to containment if the 6" HHSI line were severed. Nevertheless, since the 5" break cannot cause the complete loss of either the HHSI or of the GGP flow, a 6" break in loop 1, simulating a severed HHSI EGGS line by assuming all HHSI flow to loop 1 is lost, was analyzed.

The plots provided in the section: "COMPLETE SETS OF PLOTS FOR ADDITIONAL CALCULATIONS" (set G-1 through G- 18) show this break to be similar to the 5" break and that this 6" break is not limiting.The GGP line is 1.5", which is too small to be a limiting SBLOGA if it were severed. If this line were severed, the HHSI portion of the EGGS flow would still be available to the broken loop. Nevertheless, a 1.5" break in loop I Simulating a severed GGP EGGS line by assuming all GGP flow to the broken loop I is lost, was analyzed.

The plots provided in the section: "COMPLETE SETS OF PLOTS FOR ADDITIONAL CALCULATIONS" (set F-lI through F-18) show this 1.5" break is not limiting.

Attachment 1 to TXX-06 125 Page 9 of 216 At a June 14 phone call a question was asked regarding the elevation of the break in the cold leg for the case of the severed 6" ECCS line (Figures G-1 through G-18). The concern was that since the severed EGGS line comes in at the top of the cold leg, it might vent more steam than the model, which might vent more two-phase since it drawvs from the middle of the cold leg pipe. This would potentially help clear the broken loop seal earlier in the model than in reality.The break junction elevation in the cold leg is "centrally located" in the CPSES model. The other options are "upward oriented" and "downward oriented".

While drawing from the center of the cold leg might facilitate loop seal clearing wvith respect to an "upward oriented" break which draws from the top of the cold leg, it is also true that a break at the top of the piping would discharge more steam and consequently depressurize faster and with less loss of inventory.

In practice, drawing from the top as opposed to the center of the pipe would have no impact because the break flow is choked for the duration of the transient and the Moody model must override any upstream stratification.

In order to confirm this last observation, thle case of the severed 6" EGGS line (Figures G-1 through G-18) was re-run with the break junction option switch set to "upward oriented" from the usual "centrally located" setting. The results were virtually identical and there was no difference in PCT.e. On page 4-4 of ERX-04-004 (Ref. 3), the discussion of the 4-inch break analysis indicates results showed all of the intact loop seals cleared but the broken ioop did not. As the broken ioop is the path of least resistance, it typically clears first.Provide validation and benchmarks against integral test data that supports this condition.

TXU Power Response: While it is said on page 4-4 of Reference

[3] that "all but the broken loop seal clear (Figure 4.8)", what actually happens, as shown in the cited figure (Figure 4.8 of Reference

[3]), is that the broken loop seal clears first, followed shortly (within a minute) by the others. However, immediately after loops 2, 3 and 4 clear, the broken loop 1 seal re-plugs.

The fact that the broken loop seal stays plugged for key portions of the transient is conservative for GPSES, as explained in the next paragraphs.

When the broken loop seal clears (and stays cleared for a significant amount of time) the most direct vent path is established from the tipper plenum to the break, resulting in the fastest energy removal and depressurization rates. These factors increase EGGS injection and ultimately result in the accumulator set point being reached sooner. In addition, inventory loss is also minimized when the broken loop seal clears because the break flow has a higher void fraction and quality in that case. Last but not least, it appears that for GPSES, wvater distribution throughout the system tends to be more favorable when the broken loop seal clears. These factors tend to lower the PCT.

Attachment I to TXX-06 125 Page 10 of 216 That is wvhy it is said on page 4-4 of Reference

[3] that "all but the broken loop seal clear (Figure 4.8)" to convey the key point that this break did not benefit from the factors discussed in the previous paragraph wvhich tend to be associated with the broken loop seal being clear for an extended period of time. In other words, the purpose of the statement was to point out a conservative element of that SBLOCA calculation.

But again, the broken loop seal does clear first, as shown in Figure 4.8 of Reference

[3].During the March 29 teleconference, the NRC clarified that although they recognized that clearing the broken loop was beneficial, their concern was that it might be even more beneficial to clear all three intact loops, while the broken loop remained Plugged, as essentially occurred in the 4" break analysis of Reference

[3]( PCT183O 0 F). TXU had previously had this concern as wvell and reported during the teleconference on an analysis of a 4" break (with some inp~ut changes relative to the base case of Reference

[31) that resulted in clearing only the broken loop seal and where all the other loop seals remained plugged. The result of that analysis was a lower PCT (174 0 17). Additional infonrmation on that analysis is provided in the response to the next question (Q.7.e.i) and the corresponding plots are provided in the section: "COMPLETE SETS OF PLOTS FOR ADDITIONAL CALCULATIONS" (set H-I through H-18)As discussed in the March 29 teleconference, locating an experiment where all loops except the broken loop clear will not help the review because TXU's point is not that the loop seal clearing configuration of the 4" base case presented in Reference

[3] (also in figure set B-l through B-I8) would necessarily always occur, but rather, that for CPSES it results in a higher PCT than the case discussed in the previous paragraph (set H-I through H-18), wvhere only thle broken ioop seal cleared. The latter is the most likely loop seal clearing configuration, based on for example the S-UT-8 test. Nevertheless, the methodology has been benchmarked against at least one integral test involving loop seal phenomena.

That was the S-UT-8 test at the Semniscale Mod-2A facility.

In that benchmark, the loop seal phenomena wvere wvell predicted.

This and several other benchmarks are presented in Appendix E of the topical report which is the subject of Reference Q.7.g.(1).

i. Please provide the results of a case with only the broken loop cleared. If additional loops clear, in addition to the broken loop, provide validation and benchmarks against integral test data that supports this condition.

TXU Power Response: As mentioned above, the CPSES SBLOCA PCT tends to be lower wvhen the broken loop seal clears (and stays clear for a significant amount of time). This is verified below for the limiting 4" break presented in Reference

[3].In order to demonstrate this point, a case in which ONLY the broken loop clears was run as follows: Three changes wvere made to the 4" base case break of Reference

[3]: (1) MDAFW wvas provided to loops 2 and 3 instead of l and 4, (2) ECCS to the broken loop I was turned off and (3) the intact loop seals were lowered to make them harder to clear. This is an artificial case, since items (2) and (3) involve physical changes which are not in the plant itself but all 3 Attachment I to TXX-06 125 Page I11 of 216 changes wvere necessary in order to satisfy this RAI. The calculation results demonstrate that wvhen the broken ioop seal clears, the transient is less limiting than wvhen it does not. The PCT for this case (I1748 0 F) is slightly lower than for the base case 4" break (I1830 0 F), as illustrated by the plots provided in the section: "COMPLETE SETS OF PLOTS FOR ADDITIONAL CALCULATIONS" (set H-I through H- 18).Since not clearing the broken loop seal is conservative for CPSES, the CPSES model conservatively assigns auxiliary feedwater so as to minimize the chances of the broken loop seal clearing wvith respect to the other loops. As explained on page 4-12: "The driving force for loop seal clearing is the pressure differential between the hot leg and the cold leg. The resisting force preventing the clearing is the amount of liquid in the loop seal, the water level, etc. This resistance to clearing is not the same in the four loop seals because: Loop 1 has the break, motor-driven auxiliary feedwater (MDAFW) and turbine-driven auxiliary feedwater (TDAFW); Loops 2 and 3 have only TDAFW and Loop 4 has MDAFW, TDAFW and the pressurizer...

Thus, because they have less feed water, loops 2 and 3 condense less water on the primary side of the tubes, so there is less water and consequently a lesser resistance to clearing the loop.Similarly the No. I loop seal has more water than it would have if the MDAFW were not connected to its steam generator and that is wvhy the available MDAFW pump (I pump is taken out by single failure) is deliberately connected to loops 1 and 4: in order to give loop I (which has the break-) the least chance to clear.Another factor that reduces the chances of clearing the broken loop seal is ECCS injection into the broken loop. This provides additional water to the loop seal making it more difficult to blow.In spite of these modeling features, as shown in Figure 4.8 of Reference

[3], the broken loop seal does clear first, followed shortly (wvithin a minute) by the others wvhich, when they blowv, seemn to cause the broken seal to re-plug. Thus, the broken loop is not clear for any significant portion of the transient.

For CPSES, this results in a higher PCT than if the broken loop were clear and the others remained plugged and is therefore a conservative calculation.

Attachment 1 to TXX-06 125 Page 12 of 216 f. On page 4-9 of ERX-04-004, fluid in the hot legs and RSG was prevented from flowing back into the core. Please explain how the steam velocities can hold up water in the RSG inlet plenums following a SBLOCA in the 3-5 inch break size range. The velocities in this region, once the loop seals have cleared should be wvell below the flooding limit. Please show that the vapor velocities are sufficient to hold tip the water in the hot legs and RSG's. Does this water then enter the core and provide cooling over the long term that artificially reduces the PCT during the long term? Please explain and discuss the justification/conservatism of the model presented in the submittal.

TXU Power Response:

References:

Q.7.f.(l)

V. H. Ransom, et. al., "RELAPIMODD2 Code Manual Volume 1: Code Structures, Systemns Models and Solution Methods, NUREG/CR-43 12, August 1985.Q.7.f.(2)

Y. Taitel and A. E. Dukier, "A Model of Predicting Flow Regime Transitions in Horizontal and Near Horizontal Gas-Liquid Flow," AIChE Journal, Vol. 22, pp. 47-55, 1976.The water hold uip in the (unplugged loops) steam generator inlet plena over the time period in question (450 to 650 seconds for the 4" limiting break) occurs as a natural consequence of the flow regimes in effect. It does not occur as a result of flooding.

Typically, flooding occurs when a liquid film on a vertical wvall is prevented from flowing downward due to the velocity of the vapor. Flooding is not applicable to this situation because, the flow pattern in the steam generator inlet plenum, which is a vertical volume, is vertically stratified flowy in this time period of interest, which is not consistent with flooding.

Rather, steam velocities and mass fluxes are low when vertical stratification occurs. In order to visualize the situation consider that the steam generator inlet plenum is being fed by the hot leg, wvhich has a horizontally stratified flow pattern during this time. Thus, the horizontally stratified flowy from the hot legs enters the steam generator inlet plenum from the side. The flow remains stratified by going from horizontally stratified in the hot leg to vertically stratified in the steam generator inlet plenum, with the liquid underneath, essentially stagnant, and the vapor flowing from the top upward into the tubes. During this time period, the total mass flow rate out of the inlet plenum into the tubes is approximately the same as the mass flow rate in from the hot legs and this gives the impression of liquid being "held uip". Eventually, this liquid is scavenged out of the inlet plenum as droplets, which are entrained into the tubes flowing forward in an annular mist flow regime. As seen from the discussion above, this water does not re-enter the core.

Attachment 1 to TXX-06 125 Page 13 of 216 Regarding demonstrating that vapor velocities are sufficient to hold Lip wvater in the hot legs and in the steam generator inlet plenum it must be stipulated that they are not sufficient for flooding since flooding is not the mechanism at play. Rather, for the horizontal volumes of interest, i.e. the relevant hot legs, the steam velocities are consistent with the horizontal flow map in ANF-RELAP, as described for example in Section 3.1.3.1.2 of Reference

[Q.7.f.(l)].

This horizontal flowv map is one developed by Taitel and Dukler, Reference

[Q.7.f.(2)].

For the vertical volumes of interest, namely the relevant steam generators inlet plena, the vertical stratification criterion is given in for example in Section 3.1.3. 1.1 of Reference

[Q.7.f.(1)]

and the velocities are consistent with this pattern.Regarding explaining and justifying the conservatism of the model presented in the submittal, it is the same model already approved by the NRC in Reference

[10] with respect to the phenomenology being discussed in this question.

In fact it is the same model in all respects except for the changes addressed in Q.7.m.g. Describe how condensation of ECCS injection is modeled. Provide a reference or validation of the ECCS condensation model.TXU Power Response:

References:

Q.7.g.( 1) Letter, Ashok Thadani (USNRC) to Gary Ward (SPC), "Acceptance for Referencing Topical Report XN-NF-82-49(P), Revision 1, "EXXON NUCLEAR COMPANY Evaluation Model EXEM PWR Small Break Model", July 1988.Q.7.g.(2)

V. H. Ransom, et. al., "RELAP/MODD2 Code Manual Volume 3: Developmental Assessment Problems, EGG-SAAM-6377, April 1984.Item 2.4.2 in Reference Q.7.g.(1) reads as followvs: "2.4.2 Condensation Heat Transfer... ANF-RELAP uses the condensation heat transfer correlations from RELAP5JMOD2 without alteration.

Thus the validity of these correlations has been verified by the full extent of the RELAPS development and assessment effort (Ref.[Q.7.g.(2)J).

We find the condensation heat transfer package in the code to be acceptable." As described at the top of page 2-1 of Reference

[10] and in the NRC SER attached to it, the CPSES Small Break LOCA methodology is essentially an application of the SPC methodology of Ref.Q.7.g.(1), (also Ref. 1.1 of Reference

10) and, specifically, the condensation model is the same. Specific assessment of this condensation model is available in Sections 2.2.11 & 2.2.12 of Ref.Q.7.g.(2).

Attachment 1 to TXX-06 125 Page 14 of 216 h. Since PCT appears to be v'ery sensitive to loop seal behavior for CPSES, please show the sensitivity of ioop seal nodalization to PCT for the limiting break. Please also address the impact of only one loop seal cleared. Provide validation and benchmarks against integral test data that supports this condition.

TXU Power Response: The existing ioop seal nodalization is part of the NRC-approved methodology for the original steam generators (OSGs), which is not affected by the introduction of the replacement steam generators (RSGs). It complies wvith the guidelines for ANF-RELAP nodalization.

It does not involve a complex configuration but, rather, it is simply a "PIPE" component.

In short: there is no theoretical basis to change the CPSES loop seal nodalization.

Nevertheless, a calculation of the limiting break was performed, where the number of nodes in all loop seals wvas doubled from 4 to 8. That calculation's results are essentially the same as the base case's (PCTl84l 0 F vs.1830 0 F). A full complement of plots for that calculation is provided in the section: "COMPLETE SETS OF PLOTS FOR ADDITIONAL CALCULATIONS" (set J-1 through J- 18)Regarding the clearing of only one loop seal, such a case is being provided in connection wvith the response to Q.7.e.i above. The plots for that case, in the section: "COMPLETE SETS OF PLOTS FOR ADDITIONAL CALCULATIONS" (set H-1 through H-1 8), showed that for CPSES, the loop seal clearing configuration for the limiting 4" break presented in Reference[3] was limiting.

Benchmarks against integral test data involving loop seal phenomena are presented in Appendix E of the topical report which is the subject of Reference Q.7.g.(I).

For example, in the benchmark for the S-UT-8 test at the Semniscale Mod-2A facility the loop seal phenomena were wvell predicted by this methodology.

Although the SBLOCA PCT can be sensitive to loop seal behavior, the loop seal behavior itself seems stable for a given calculation.

Recall, that in order to obtain the case presented in Q.7.e.i above, which cleared only one loop seal, it was necessary to: (a) switch the loops receiving MDAFW, (b) eliminate ECCS to the broken loop and (c) bias the intact loop seals to prevent them from clearing.

All this indicates the current loop seal clearing pattern for the limiting 4" break is stable. The nodalization study discussed in the first paragraph shows the same thing, since that run was nearly identical to the base case. The time step study and the core cross-flow resistance study also test the stability of the loop seal clearing pattern and they too confirm that the loop seals clearing pattern for the limiting 4" break is stable.Based on the above, the current loop seal clearing configuration for the limiting 4" break is stable, not affected by loop seal nodalization and, in any case, is more limiting than the benchmarked loop seal clearing configuration, as discussed in Q 7.e and Q.7.e.i.At a June 14 phone call the NRC said they would have preferred to see more nodes in the vertical portion of the volume immediately upstream of the pumps for the loop seal nodalization study discussed above. Howvever, since they were not specific in the original request and TXU did double the number of nodes there, wvith the results showing almost no sensitivity, the NRC agreed there are no action items here. An additional request was made that TXU Would make sure there wvas no SLUG regime flow in the ioop seals during loop seal clearing.

There was no slug regime flow in the ioop seals at any time. Figure 2 in the"MISCELLANEOUS FIGURES" section of this document shows the only flow regime types to be: bubbly, stratified and mist, depending on the loop seal and time in the transient.

Attachment 1 to TXX-06 125 Page 15 of 216 i. In the cross flow sensitivity study please identify the magnitude of the cross flow resistance for each case and describe/justify the method for calculating cross flow resistance.

TXU Power Response: As stated in Section 3.2.2 of Reference

[10], three values are used in the crossflow resistance study: nominal, 10 times nominal, and nominal divided by 10. As stated in the TER of Reference

1. 1 of Reference

[10], the standard methodology implemented in COBRA analyses is used for calculating nominal cross flowv resistance.

This methodology is based onl textbook relationships for flow across tube banks and is provided in one of the two attachments to Reference Q.7.l(l), pages 3-1 and 3-2. At a June 14 phone call the NRC asked what were References (2) and (3) mentioned in those pages. References (2) and (3) of pages 3-1 and 3-2 of one of the two attachments to Reference Q.71l(l) are pages 333 and 339, respectively of the same book: Knudsen, James, G. and Katz, Donald, L., Fluid Dynamics and Heat Transfer, McGraw-Hill Book Company, New York-, 1958.j. Demonstrate that above the two-phase level no liquid is entrained in the steam leaving the two-phase surface via the drag model in RELAP5 that would artificially dc-superheat the steam at the hot spot for the limiting SBLOCA.TXU Power Response: Item 2.2.3 of Reference Q.7.g.(1) reads in part: "12.2.3 Mixture Level Model... This [ANF-RELAPJ model corresponds more closely with the physical situation for the presence of a mixture model than the model found in RELAP5IMOD2.

In Appendix C of Reference 4 [Rcf.Q.7.g.(2)J, ANF documents the use of the code to calculate the results of the ORNL THTF level swell test (discussed in Section 2.3.3).Because of the good agreement between ANF [-RELAPI code calculations and THTF test data, the staff concludes that this code modification is acceptable for licensing analysis.As described at the top of page 2-1 of Reference

[10] and in the NRC SER attached to it, the CPSES Small Break LOCA methodology is essentially an application of the SPC methodology of Ref.Q.7.g.(1), (also Ref. 1. 1 of Reference

[10]) and the mixture level model is the same.Also, the additional figures requested in Q7.a.iii include the steam temperatures adjacent to the hottest clad temperature nodes. Plots of those temperatures (Figures 15 of the A through J sets, i.e. A-15, B-15... J-15) provided in the section: "COMPLETE SETS OF PLOTS FOR ADDITIONAL CALCULATIONS" show the steam is highly superheated with temperatures following those of the adjacent cladding.

Attachment 1 to TXX-06 125 Page 16 of 216 k. For the 3-inch break, please explain why there is a 1700-second period of cooling after the initial heat up at 500 to 1000 seconds for the 3-inch break?TXU Power Response: The second period of cooling for the 3" break lasts approximately 700 not 1700 seconds. As discussed in the March 29 teleconference with the NRC, TXU assumes this is a typo and wvill answer the question accordingly.

The initial heat uip that began at 500 seconds is arrested around 900 seconds (Figure 4.19 of Reference

[3]) wvhen loop seal clearing begins to occur.As thle loop seals clear, water in the ioops is free to flow back into the core. When enough of this water gets there the rods undergo the preliminary quench that occurs near 1000 seconds.Figure 4.18 of Reference

[3] showvs the core collapsed level increase at this time, wvhich results from this effect. The loop seal clearing allowvs the RCS pressure to drop more rapidly as shown in Figure 4.13 of Reference

[3]. This in turn causes an increase in ECCS flow as shown in Figure 4.23 of Reference

[3]. The additional wvater from the loops in combination with the increased E 'CCS flow slows down the inventory loss so that it takes from approximately 1000 seconds to approximately 1650 seconds for the collapsed core level to drop below its mid-point elevation (6 ft. Figure 4.18), which typically corresponds to the onset of critical heat flux shown to occur around that time (1650 seconds, Figure 4.19). This second heat-up typically is the main heat-uip for CPSES, as discussed in the next question (Q.7.1).1. The temperature profiles for CPSES, in the submittal, do not resemble SBLOCA PCTs for plants of this class and power level. Please explain and discuss why CPSES is unique.TXU Power Response:

Reference:

Q.71l(1) R.A. Copeland (SPC) to Frank Orr (USNRC), "Response to NRC Concerns about SPC SBLOCA Model," March 17, 1994, RAC:94:037.

At the March 29 teleconference, the NRC agreed that lack of resemblance to "SBLOCA PCTs for plants of this class and power level" refers to the preliminary heat up that is seen prior to loop seal clearing for the RSG cases, an example of which would be the heat up between 300 and 500 seconds seen in Figure 4.7 of Reference

[3].While this pre-heat uip makes the PCT temperature histories for the RSG appear unusual, it is not unique: it also appears, although less conspicuously, in other PCT histories.

For example, in the PCT history for the sample application provided on page 31 of an attachment to Reference Q.71l(1).

Other examples are the CPSES clad temperature histories for the OSGs, which use NRC-approved methodology and which can be seen in Figures 3.9 and 3.22 Reference

[10].In these examples, the "pre-heatups" appear as "blips" rather than "humps" but they are all due to wvater being temporarily held away from the core, e.g. in the loops, eventually finding its way back into the core when loop seals clear. The calculations wvhere these "pre-heatups" are absent or small simply reflect that a larger fraction of the RCS inventory stays in the reactor vessel for the duration of the transient as opposed being temporarily held in the loops.In either case, i.e. whether this pre-heatup occurs or not, the actual heat uip is similar because it either happens after water held in the loops returns to the vessel and is eventually lost via Attachment 1 to TXX-06125 Page 17 of 2l6 break or, if that water remained in the core, after it is lost to the break. Thus, the "pre-heatup" is not the definitive heat up, i.e. the PCT doesn't occur in this phase because water held in the loops returns to the vessel when loop seals clear. The pre-heatup is more pronounced in the RSGs than in the OSGs because there is more water held in the loops in the RSGs due to their much larger steam generator tube Volume and hence appears in the temperature profiles as an unfamiliar "hump", rather than the often ignored "blip". Note the RSGs are a A76 model, wvith an atypically large steam generator number of tubes! tube-side volume (5532 vs. 4578/1303 ft3 VS. 967 ft 3 , see Table 2.1 of Reference

[3]).At a June 14 phone call the NRC asked what causes the dip in clad temperature around 900 seconds seen for example in Figure B-7, corresponding to the clad temperatures for the limiting 4inch break base case. This inflection is due to water redistribution between the Hot Assembly, Central Region and Average Core. Figures B-2, B-3 and B-4, respectively, show void fractions at three core elevations:

bottom (nodes, 112, 130,150), lowvest a = 1 elevation (nodes, 122, 140, 160) and an intermediate elevation (nodes, 117, 135, 155) in each of these three core regions. At the time period of interest, around 900 seconds, thle topmost elevation (nodes, 122, 140, 160) remain at a = 1. This shows that no water has entered either of these three core regions from the top. Further evidence of this is provided in Figure B-5 that shows no liquid at this time in the upper plenum either. Similarly, there is no significant trend change in the void fractions in the lower elevations (nodes, 112, 130,150).

Additional inspection of mass flow rates from the lowver plenum to the various core regions (flows from volume 111 to 112, 130 and 150) also show no change over the period of interest.

These observations indicate there was no change in the amount of water entering these three core regions either from above or from below. A look at the void fraction behavior at the intermediate elevation (nodes, 117, 135, 155) however, is revealing.

While similar oscillations in void fraction are seen at that elevation in all three regions, the highest void fractions Occur in the average core region, then in the central region and finally in the hot assembly.

This hierarchy indicates net water movement away from the larger regions towvards the hlot assembly at intermediate elevations.

A look at representative cross flows during the period of interest shows a step reduction in the outflow from the hot assembly.

Figure B-15 shows a reduction in the hot channel superheat about the PCT elevation which is consistent with more of the upstream flow remaining in the hot assembly.

Thus, these observations indicate that the inflection in temperature is due to a redistribution of flow between the various core regions, where outflow from the hot assembly region is temporarily reduced to satisfy energy and momentum conservation equations, most likely compensating for an earlier overshoot.

Note that the cross flow resistance study (Figure 4.37 of Reference

3) identifies the limiting cross flow resistance.

Heat transfer and flow regime flags in the core were examined but no significant changes occurred during the time period of interest that could account for the inflection in the clad temperature history.It should be noted that this inflection in the clad temperature is not unique to CPSES, nor to TXU's application of the SPC methodology.

For example, a similar inflection appears in the ruptured node clad temperature history for the sample application provided on page 31 of an attachment to Reference Q.71l(1) (mentioned above).

Attachment 1 to TXX-06 125 Page 18 of 216 Also at the June 14 phone call the NRC asked TXU to investigate, with respect to the Appendix K lockout rule, the apparent return to nucleate boiling associated with the quench occurring due to loop seal clearing around 450 seconds in Figure B-7 showving clad temperatures for the limiting 4 inch break base case. This issue was previously addressed in item 3.0 of Reference Q.7.g.(1) which reads as follows: "43.0 COMPLIANCE WITH NRC REQUIREMENTS

..the post-CHF heat transfer requirements of Appendix K specify return to nucleate boiling be locked out once CHF has occurred during blowdown.

This requirement is not appropriate for SBLOCA." m. In addition to the feedring RSG changes, the licensee is proposing two changes to its reactor vessel nodalization in the ANF-RELAP model for CPSES Unit 1. First, four upper downcomer nodes are collapsed into two nodes. Nodes 104 and 106 are being combined and nodes 100 and 102 are being combined.

Second, 11... .the flow area between the upper downcomer and upper head "spray holes" is being updated to reflect more accurate, recently developed design information." These changes are only being proposed for the CPSES Unit 1 ANF-RELAP model.i. Explain what evidence provides reasonable assurance these changes will accurately reproduce conditions at CPSES Unit 1.TXU Power Response: Both downcomer nodalizations (proposed and current) are essentially the same over more than 4 1 5 t1h, of the downcomer length, and after that both connect to single lower plenum node.Thus, these nodalizations are essentially the same because, even in the current nodalization, the downcomer is not azimuthally split for its entire length, rather, it becomes one node azimuthally just beneath the cold leg elevation.

The only reason for the change is to provide a more uniform pressure boundary condition for the loop seals during loop seal clearing, as described in Q.7.m.ii.(1), (2) and Q.7.m.iii. (Note that b 'ased on the annular geometry of the region alone it is reasonable to expect an azimuthally uniform pressure distribution in the dowvncomer.)

Conditions at CPSES Unit 1 are currently modeled either wvay. Both configurations have an extensive track record. For example, the azimuthally split configuration is used for the small break LOCA in ANF-RELAP with the D-4 steam generators and has been approved by the NRC for use in that manner (Reference

[10]).By contrast, the single azimuthally configuration proposed here for the RSGs, was used in all the applicable benchmarks presented in the topical report which is the subject of Reference Q.7.g.(l).

For example, Figure E.4.lin that topical report shows that both cold legs in the S-UT-8 test Semiscale Mod-2A model feed into a single downcomer node. The same is true for the LOFT L3-1 test as shown in Figure B.4.1 of the same topical report.

Attachment 1 to TXX-06 125 Page 19 of 216 The change proposed to the flow area between the uipper downcomer and uipper head "spray holes" is an update to reflect more accurate design information that was obtained as a result of an unrelated activity.

Additional information on this change is provided in Q.7.m.vi, and Q.7.i-f.vii.

The rationale here is that the most accurate design information available should be used on general principle.

ii. The licensee states that collapsing the four downcomer nodes into two nodes " ... makes the model more robust numerically for A76 applications, but does not significantly improve the numerics of the Unit 2 model." (1) Why did the ANF-RELAP model, as modified for A76 steam generators, need to have its 'numeric robustness' improved?TXU Power Response: The methodology demonstration submitted via Reference

[3] includes tw~o sensitivity studies that are essentially tests for numerical robustness:

the crossflow sensitivity study and the time step sensitivity study. Prior to consolidating the two sets of nodes that azimuthally split the upper 1 1 5 1h of the downeomer in two, the time step sensitivity (or "numeric robustness")

for key A76 calculations was less than adequate.

The proposed consolidation resolved the problem.(2) Explain why collapsing the four downcomer nodes into two nodes affects the 'numeric robustness' of the A76 applications but not the D4 applications.

TXU Power Response: The D-4 applications met numerical robustness tests including the crossflow sensitivity study and the time step sensitivity study, with the original nodalization and therefore did not require change. The A76 did not until the nodes were collapsed, as explained above in Q7.m.ii.(l).

The driving force for loop seal clearing is the pressure differential between the hot leg and the cold leg. The resisting force preventing the clearing is provided by the liquid in the loop seal, e.g. the wvater "plug" resistance.

Thus, the mechanism for loop seal clearing involves a threshold, it either happens or doesn't. It is not a matter of degree and does not take place gradually.

As a result of this threshold effect, differences in initial and/or boundary conditions may result in a different loop seal clearing sequence.

This in and of itself may not impact the transient much, if similar loop seals are cleared for similar time periods. However, in the A76 prior to the proposed collapsing of the 2 downcomer nodes, the minor differences that occur in the clearing sequence may have been amplified by the fact that, once a loop seal cleared, the driving pressure differential for loop seal clearing was affected by the water that found its way back into the dowvncomer after the clearing.

This effect might be more significant in the A76 than in the D-4 because of the followving:

Attachment 1 to TXX-06 125 Page 20 of 216 Prior to collapsing nodes, 2 loops see slightly different boundary pressures in the downcomer than the other 2 loops. After the nodes are collapsed, all four loops see the same boundary condition (pressure) in the downcomer.

Thus, after the change, if one loop blows, that wvater affects all the other loops equally. However, prior to the change, one loop blowing affects the pressure more in the loop that it connects to via downcomer node. In the case of the A76, the large SG tube volume results in a correspondingly larger wvater volume being blown into the downcomer at the time of loop seal clearing.

Thus, an otherwise minor clearing sequence difference associated with, say a time step study, now changes which loop clears next and which loops remain clear. Since the D-4 has a smaller SG tube volume, the slug associated with blowing a loop seal is correspondingly smaller and apparently does not affect what loop seals subsequently clear, even if loops attached to different downcomer nodes clear in a slightly different sequence.It is also possible that for the D-4, the mninor differences in the amount of liquid in the loops associated with the various sensitivity studies, in combination with the small variations in the driving pressure differential did not affect the loop seal clearing pattern because the liquid amounts were small and thus their relatively magnitudes were significantly different from each other, so that the clearing pressure thresholds wvere also significantly different from each other. Thus, fluctuations in magnitude did not affect hierarchy.

'Conversely, for the A76 with the larger volume, the resistance to loop seal clearing increased and became relatively similar among the loops. In this case, the clearing pressure thresholds also become similar in all loops and then minor variations in either one have a big impact on the loop seal clearing sequence itself.iii. Nodes 104 and 106 are being collapsed into a single node. Nodes 104 and 106 each had two RCS cold leg inputs. Collapsing them into a single node will have all four RCS cold leg inputs being routed to a single node. Explain the impact of having all four RCS cold leg inputs being routed to a single node, include the rationale for the original model configuration.

TXU Power Response: Both downcomer nodalizations (current and proposed) are essentially the same. They are exactly the same over more than 4 1 5 ths of the downcomer length and prior to entering the single lower plenum node. This is because for both nodalizations the dowvncomer is not azimuthally split for its entire length, rather, it becomes one node azimuthally just beneath the cold leg elevation.

The impact of collapsing the nodes is to provide a more uniform pressure boundary condition for the loop seals during loop seal clearing, as described in Q.7.m.ii.(l), (2) and Q.7.m.iii.(Note that based on the annular geometry of the region alone it is reasonable to expect an azimuthally unifonri pressure distribution in the dowvncomner.)

Attachment 1 to TXX-06 125 Page 21 of 216 The rationale for the original nodalization was that part of the connection between the nodes is partially blocked by hot legs in the region. However, the presence of the hot legs does not prevent the pressure in the upper downcomer region from being relatively uniform. Having all cold legs see the same pressure boundary condition, as previously explained, is the goal and therefore the variable of interest is the pressure.

Still, another important consideration is that the reason the nodes were split in the original model was to make it more difficult for intact loop EGGS to find its way to the break. Thus, from the standpoint of thle rationale for the original model, the proposed change is in the conservative direction.

iv. NNherc is the break location relative to the loop connections to the downcomer for the new model? Please provide the reference or sensitivity studies preformed to justify this unique original nodalization prior to this requested change (what is the impact of this change on the limiting SBLOCA PCT)?TXU Power Response: Based on the March 29 conference call, it appears the NRC's concern wvith "this unique original nodalization prior to this requested change" is that the break junction is connected to the vessel side as opposed to the pump side of the break. Alternatively, that the break junction is connected to the inlet Of Volume 495 where the NRC was concerned it might be more limiting to connect it to the Outlet Of Volume 490 (see Figure 2.3 of Reference

[10]).The reason why the connections are as they are is that RELAP5/MOD2 guidance is not to connect more than one process model to the same volume. Since EGGS and BREAK FLOW are both process models and given that EGGS is connected to Volume 490, then the BREAK FLOW is, logically, connected to the closest location to the break in the adjacent volume 495.It should be noted this nodalization is NOT unique to TXU and that the same configuration was used by SPC, e.g. Figure 4.1 of the topical report of reference Q.7.g.(l).

As far we are aware, Framatome still uses this configuration.

With respect to the TXU application of the SPG model, the break location has not changed either. The break location relative to the loop connections is shown in Figure 2.3 of Reference'

[10]. This break location is part of an approved model as per SER dated 9/4/1996 and attached to Reference

[10].(Several years ago, in mid 1997, TXU performed a sensitivity study on this issue. It found that the NRC-approved configuration, i.e., that of Figure 2.3 of Reference

[10], was more limiting)

Attachment 1 to TXX-06 125 Page 22 of 216 (1) Are the hot leg nozzle gaps modeled in the SBLOCA analysis?

Please explain and justify this modeling decision.Please describe and justify all of the hot side leakage paths modeled in the SBLOCA analyses.TXU Power Response: In addition to the "spray holes", discussed in Q.7.m.vi and Q.7.m.vii, the hot leg gaps are the only hot side leakage paths included in the A76 model. These leakage paths are modeled as shown in Figuire 2.2 of Reference

[10]. The justification for including these in the model is that they are there, i.e. that the most accurate design information available is used on general principle.

Based on the March 29 conference call, it appears the NRC has a concern that these gaps may change during the transient due to thermal expansion differentials.

The area of each CPSES nozzle gap is very small 0.0038 ft 2 .Nevertheless, in order to address this concern the NRC accepted TXU's proposal to analyze the limiting 4" break without the hot leg gaps. That calculation's results are slightly higher (approximately 50 0 F) than the base case's, although the actual transient development is essentially the same. A full complement of plots for that calculation is provided in the section: "COMPLETE SETS OF PLOTS FOR ADDITIONAL CALCULATIONS" (set I- I through 1- 18).N. Where is the break location relative to the loop connections to the downcomer for the new model?TXU Power Response: Break location was discussed in the answer to the previous question, Q.7.m.iv.

Again, the break location has not changed with respect to that in the NRC-approved model, shown in Figure 2.5 of Reference

[10]. The NRC SER for that model is attached to Reference

[10].vi. What is the ". ..more accurate, recently developed design information" used to set the revised flow area between the upper downeomer and upper head "spray holes" and how is current value more conservative?

TXU Power Response:

Reference:

Q.7.m.vi.

Westinghouse Letter "...Reactor Vessel Upper Head Region Bulk Fluid Temperature Design Basis," WPT-16476, October 8, 2003 (VL-04-002358).

The more accurate, recently developed design information was obtained as a result of an unrelated activity.

It is Reference Q.7.m.vi.

Attachment I to TXX-06 125 Page 23 of 216 The current value is more conservative because a preliminary calculation was performed for CPSES-2 with the Current, NRC-approved, Reference

[10] model, wvhere only this parameter was changed to the value given in the above reference and the resulting PCT was lower.However, the fact that the value of this parameter is not being updated for CPSES-2 at this time is not the subject of thle present LAR.vii. What is the flow area of the upper head spray nozzles, for CPSES Unit 1 and Unit 2?TXU Power Response: Approximately 28 square inches, per Reference Q.7.m.vi above.n. Explain how the increased size of the A76 tube bundle is apportioned between the existing nodes. Confirm that the increased size of the nodes does not violate any underlying assumptions of the models such as length or bend radii.TXU Power Response:

Reference:

Q.7.n Siemens Power Corporation, "Guidelines for PWR Safety Analysis:

Small Break LOCA Analysis (SRP 15.6.65) ," EMF-2062 (P), Rev. 0, June 17, 1998.The tube bundle node dimensions in the ANF-RELAP model are defined according to the methodology guidelines of Reference Q.7.n. The tube bundle region is represented by a total of 8 vertical nodes; 4 oriented vertically upward and the remaining 4 oriented vertically downward.

The top most nodes depict the U bend and are connected by a horizontal junction.This nodalization scheme is the same one used with the D-4 steam generator and is shown in Figure 2.3 of Reference

10. The ANF-RELAP tube bundle node dimensions for both the A76 and the D-4 steam generators are given belowv in Table 1. Thle two steam generators are shown in Figures 2.1 and 2.3 of Reference 3.The relatively higher number of U-tubes in the A76 SG is reflected in the model via the larger flow area of 11.97 ft 2 compared to 10.46 ft 2 for the D-4 SG.The average A76 U-tube length of 34.98 ft is distributed in the tube region model with 4 up-flow and 4 down-flowv nodes as shown in Table 1, which is the same modeling approach used to represent the 28 ft average tube length of the D-4 SG.The A76 SG total U-tube volume7 amounts to 837.7 ft 3 compared to the D-4 U-tube volume of 585.78 ft 3 , and is apportioned according to the flow area and node lengths of Table 1.The A76 nodes were verified to remain within the parameters of applicability of all conservation and constitutive equations (correlations).

Attachment I to TXX-06 125 Page 24 of 216 Table 1 -ANF- RELAP SG U-Tube Nodalization D-4 SG A-76 Length (ft) _______________________

Node 1 9.06 10.41 Node 2 7.25 10.41 Node 3 7.25 10.41 Nodes 4& 5 4.44 3.77 Nodes 6& 7 7.25 10.41 Node 8 9.06 10.41 Volume (ft 3)Node 1 94.77 124.59 Node 2 75.84 124.59 Node 3 75.84 124.59 Nodes 4 & 5 46.44 45.08 Nodes 6 &7 75.84 124.59 Node 8 94.77 124.59 Flaw Area (ft 2)_____________

Nodes 1- 8 10.46 11.97 Orientation

___________

_____________

Nodes 1 -4 1+90 +90 Nodes 5 -8 1-90 -90 o. The licensee currently uses an ANF-RELAP model for its SBLOCA analysis, TXU Electric's RXE-95-OO1-P-A, "Small Break Loss of Coolant Accident Analysis Methodology," (Ref. 10). The licensee is proposing using the same nodalization for modeling the feedring RSG as is used for the preheater OSG, figure 2.3 of RXE-95-001-P-A.

Figure 2.3 of RXE-95-OO1-P-A is a single loop diagram, please confirm that the four loop model represented by figure 2.2 of RXE-95.-O0l-P-A is being used for the SBLOCA analysis.TXU Power Response: That is correct, the four loop model represented by Figure 2.2 of Reference

[10] is being used for the SBLOCA analysis.

Note that although Figure 2.3 of Reference

[10] only shows loop 1, the broken loop, in order to allow for larger digits, that figure is a partial viewv of the 4-ioop model, which is displayed in its entirety on Figure 2.2 of Reference

[10].

Attachment 1 to TXX-06125 Page 25 of 216 8. With respect to the changes in CPSES's large break loss of coolant accident (LBLOCA) methodology described in ERX-04-004, "Replacement Steam Generator Supplement to TXU Power's Large and Small Break Loss of Coolant Accident Analysis Methodologies," (Ref. 3), provide the following information.

a. In the LBLOCA analysis, containment pressure can affect PCT. As such, no failure can result in the maximum spillage to the containment which can reduce containment pressure during the late reflood peak. Please provide a reference for the analysis of the no failure condition or provide an analysis to show that the containment minimum pressure analysis with no ECCS failure does not produce the nmost limiting PCT.TXU Power Response: As explained in page 6-2 of Reference

[3], the single failure considered in the large break LOCA analysis is the failure of 1 train of RHR, which is more limiting than the loss of 1 full train of ECCS. This is because the former results in the minimum containment pressure due to the actuation of 2 trains of containment spray, whereas in the latter case, only 1 train of containment spray injects. In order for the limiting single failure, namely the loss of I train of RHR, to result in a lower PCT than the case with no failure at all, it would be necessary for the incremental break flow resulting from a lower containment pressure associated with the additional condensation from the spillage of 2 rather than 1 train of RH-R to be greater than the increment in make Lip flow due to the injection from 2 rather than 1 RHR train. Simple engineering judgment says this cannot be the case. Given that both scenarios: "loss of 1 RHR train," as well as "no failure" have 2 trains of containment spray running, the increment in break flow associated with increased condensation from the additional spillage of RHR is a second order effect and could not possibly offset the incremental RHR flow being injected in the no failure case, which has a direct or first order effect on flooding rate and consequently on the PCT.

Attachment I to TXX-06 125 Page 26 of 216 b. What are the effects of downcomer boiling on PCT for the limiting LBLOCA?Show that the limiting LBLOCA considers this condition.

What is the worst single failure condition when downcomer boiling is a consideration?

Describe and justify the analysis that lead to these conclusions.

TXU Powver Response:

Reference:

Q.8.b H.C. da Silva, P. Salim and W. G. Choe, "Effect of Downcomer Boiling on LOCA PCT for a 4-Loop PWR with a Large Dry Containment," I 01h International Topical Meeting on Reactor Thermal Hydraulics (NUTHETH 10)Seoul, Korea, October 5-9, 2003.According to the Reference

[12] SER, item 3.3.2, TXU committed (Commitment Number 27266) to investigate this issue. According to the 2002 annual PCT report dated Dec 17, 2002, TXU communicated to the NRC that the impact of downcomer boiling was investigated within the corrective action program (SMF-2002-002036) and concluded that a PCT penalty was not warranted.

The bases for that conclusion are described in Reference Q.8.b.Thus, this is a generic issue that has already been addressed, as indicated in the previous paragraph.

Also, since the current model has already been approved for use with the OSGs (Reference

[10]) and since there is nothing about the RSGs that impacts downeomer boiling, this issue appears unrelated the present LAR. Finally, the RSGs have a lowver primary side resistance which would have a beneficial effect on the core flooding rate and core liquid region inventory, which would be beneficial in regards to this issue.c. Replacement of the RSG's could represent an increased resistance in the loop during long term cooling following an LOCA due to the larger RSG's. Please show that the RSG's do not pose a more limiting condition for post-LOCA long term cooling performance such that the boric acid precipitation time does not decrease.

A decrease in the timing to boric acid precipitation could result in less time for the operators to switch to simultaneous injection to control boric acid precipitation.

Please describe the methods to evaluate boric acid precipitation and discuss the impact of the replacement steam generators on the timing (and hence EOP guidance) to switch to simultaneous injection.

TXU Power Response: On a first order basis, post-LOCA long term cooling will be not impacted by the RSGs, for the following reason: After an LBLOCA, the RCS depressurizes rapidly to near containment pressure and reflood and core quench occur early relative to the time required for the core boric acid concentration to approach its solubility limit. After reflood and core quench the system is in quasi equilibrium.

It is during this period that the boric acid concentration begins to increase to significant levels. The RSGs would have no significant effect on the boric acid concentration during this quasi-equilibrium period because the rate of boric acid accumulation is dependent primarily on the boron concentration of the injected SI and the steaming rate in the core.

Attachment I to TXX-06 125 Page 27 of 216 There would be a second order beneficial effect on post-LOCA long term cooling from the larger primary side volume of the RSGs. This would due to the pre-LOCA RCS inventory, which is a dilution source for the sump boric acid solution.

The increased RSG primary side volume would slightly decrease the sump boric acid concentration and therefore Would decrease the rate at which boric acid accumulates in the core region. The RSGs even at high tube plugging levels would have a significantly higher primary side volume wvhen compared to the OSGs at no plugging.Finally, the RSGs have a lower primary side resistance which would have a beneficial effect on the core flooding rate and core liquid region inventory.

Increased core liquid region inventory would decrease the boric acid concentration rate of increase in the core.Regarding the TXU methodology for post-LOCA long term cooling, this part of 10 CFR 50.46 has never been submitted for reviewv before and is not currently described or referenced in the FSAR. Consequently it is not an issue for this LAR. However, TXU participates in a WOG program that is currently developing a comprehensive methodology for post-LOCA long term cooling.d. The licensee's current LBLOCA methodology is described in ERX-2000-002-P-A, "Revised Large Break Loss of Coolant Accident Analysis Methodology," (Ref. 11). Figures 2.2 and 2.5 describe the nodalization of the RCS, including steam generators.

The licensee states that, "Suffice it to reiterate that Figures 2.2 and 2.5 remain unchanged, along with the entire nodalization of the LBLOCA methodology and simply the nodal geometrical information is changed to reflect the A76 rather than the D-4 steam generators." Explain what evidence provides reasonable assurance that merely revising the nodal geometrical information is sufficient to ensure the revised model will accurately reproduce A76 steam generator conditions at CPSES Unit 1.TXU Power Response: The A76 nodes wvere verified to remain within the parameters of applicability of all conservation and constitutive equations (correlations).

Revising the nodal geometrical information is simply another way of stating that the same nodalization applied to the OSG is applied to the RSG. In any case, due to the nature of the transient, the steam generator nodalization has negligible, if any, impact on the large break LOCA. The RCS depletion is simply too fast and the secondary is thermally decoupled from the primary almost immediately after the break.

Attachment 1 to TXX-06125 Page 28 of 216 Ill. NRC RAI REFERENCES

1. TXU Power, letter dated February 17, 2005 from Mike Blevins, Senior Vice President& Chief Nuclear Officer to USNRC, re: Comanche Peak Steam Electric Station (CPSES), Docket No. 50-445, Request for reviewv of Previously Submitted Licensee Topical Reports 2. TXU Power, letter dated January 25, 2005 from Mike Blevins, Senior Vice President

&Chief Nuclear Officer to USNRC, re: "Comanche Peak Steamn Electric Station (CPSES), Docket No. 50-445, Submittal of TXU Power's Application of Non-LOCA Transient Analysis Methodologies to a Feed Ring Steam Generator Design., Topical Report #ERX-04-005, revision 0." 3. TXU Power, letter dated January 25, 2005 from Mike Blevins, Senior Vice President

&Chief Nuclear Officer to USNRC, re: "Comanche Peak Steam Electric Station (CPSES), Docket No. 50-445, Submittal of Supplement to the CPSES Loss of Coolant Accident (LOCA) Analysis Methodologies

-Topical Report #ERX-04-004, revision 0." 4. RXE-9 1-001 -A, "Transient Analysis Methods for Comanche Peak Steam Electric Station Licensing Applications," October 1993.5. RXE-9 1-005-A, "Methodology for Reactor Core Response to Steamline Break Events," February 1994.6. RXE-94-001 -A, "Safety Analysis of Postulated Inadvertent Boron Dilution Event in Modes 3, 4, and 5," February 1994..7. RXE-9 1-002-A, "Reactivity Anomaly Events Methodology," October 1993.8. GL 83-1 1, Licensee Qualification for Performing Safety Analysis in Support of Licensing Actions.9. WCAP-14882-P-A, "RETRAN-02 Modeling and Qualification for Westinghouse Pressurized Water Reactor Non-LOCA Safety Analysis." 10. TXU Electric's RXE-95-00 1-P-A, "Small Break Loss of Coolant Accident Analysis Methodology," September 1996.11. ERX-2000-002-P-A, "Revised Large Break Loss of Coolant Accident Analysis Methodology," March 2000.12. ERX-200 1-005-N-P. "ZIRLO T 11 1 Cladding and Boron Coating Models for TXU Energy's Loss of Coolant Accident Analysis Methodologies," September 2002.

Attachment 1 to TXX-06125 Page 29 of 216 IV. COMPLETE SETS OF PLOTS FOR ADDITIONAL CALCULATIONS This section contains "complete sets" of plots for the 3", 4" and 5" breaks as wvell as for all the additional cases run in connection with this RAI. Each "complete set" of plots is comprised of the variables that were plotted for the submittal (Reference

[3]) as wvell as the additional variables requested in Q.7.a. Specifically, the plots provided in this section are: A-I Primary and Secondary System Pressures B 3-in Break A-2 Hot Assembly Region Void Fractions B 3-in Break A-3 Central Core Region Void Fractions B 3-in Break A-4 Average Core Region Void Fractions B 3-in Break A-5 Upper Plenumn Liquid Fraction B 3-in Break A-6 Hot Assembly Collapsed Water Level B 3-in Break A-7 Hot Assembly Clad Temperatures B 3-in Break A-8 Loop Seal Void Fractions B 3-in Break A-9 Accumulator Mass Flow Rates B 3-in Break A-10 Break Flow Rate B 3-in Break A-I 1 Total Pumnped ECCS Flow Rate B 3-in Break A-12 TOODEE2 Clad Temperature B 3-in Break A-13 Core Mixture Level B 3-in Break A-14 Downcomer Liquid Level B 3-in Break A-15 Hot Assembly Steam Temperatures B 3-in Break A-16 Hot Assembly Heat Transfer Coefficients B 3-in Break A-17 Condensation Rate in Cold Leg Discharge B 3-in Break A-18 Break Quality B 3-in Break B-1 Primary and Secondary System Pressures B 4-in Break B-2 Hot Assembly Region Void Fractions B 4-in Break B-3 Central Core Region Void Fractions B 4-in Break B-4 Average Core Region Void Fractions B 4-in Break B-5 Upper Plenum Liquid Fraction B 4-in Break B-6 Hot Assembly Collapsed Water Level B 4-in Break B-7 Hot Assembly Clad Temperatures B 4-in Break B-8 Loop Seal Void Fractions B 4-in Break B-9 Accumulator Mass Flow Rates B 4-in Break B-10 Break Flow Rate B 4-in Break B-1 I Total Pumped ECCS Flow Rate B 4-in Break B-12 TOODEE2 Clad Temperature B 4-in Break B-13 Core Mixture Level B 4-in Break B-14 Downcomer Liquid Level B 4-in Break B-I5 Hot Assembly Steam Temperatures B 4-in Break B-16 Hot Assembly Heat Transfer Coefficients B 4-in Break B-17 Condensation Rate in Cold Leg Discharge B 4-in Break B-18 Break Quality B 4-in Break Attachment I to TXX-06125 Page 30 of 216 C-I Primary and Secondary System Pressures B 5-in Break C-2 Hot Assembly Region Void Fractions B 5-in Break C-3 Central Core Region Void Fractions B 5-in Break C-4 Average Core Region Void Fractions B 5-in Break C-5 Upper Plenum Liquid Fraction B 5-in Break C-6 Hot Assembly Collapsed Water Level B 5-in Break C-7 Hot Assembly Clad Temperatures B 5-in Break C-8 Loop Seal Void Fractions B 5-in Break C-9 Accumulator Mass Flow Rates B 5-in Break C-10 Break Flow Rate B 5-in Break C-I I Total Pumped ECCS Flow Rate B 5-in Break C-12 TOODEE2 Clad Temperature B 5-in Break C-13 Core Mixture Level B S-in Break C- 14 Downcomer Liquid Level B 5-in Break C-15 Hot Assembly Steam Temperatures B S-in Break C-16 Hot Assembly Heat Transfer Coefficients B S-in Break C-17 Condensation Rate in Cold Leg Discharge B S-in Break C-18 Break Quality B 5-in Break D-1 Primary and Secondary System Pressures B 4.222-in Break D-2 Hot Assembly Region Void Fractions B 4.222-in Break D-3 Central Core Region Void Fractions B 4.222-in Break D-4 Average Core Region Void Fractions B 4.222-in Break D-5 Upper Plenum Liquid Fraction B 4.222-in Break D-6 Hot Assembly Collapsed Water Level B 4.222-in Break D-7 Hot Assembly Clad Temperatures B 4.222-in Break D-8 Loop Seal Void Fractions B 4.222-in Break D-9 Accumulator Mass Flow Rates B 4.222-in Break D-10 Break Flow Rate B 4.222-in Break D-l1 Total Pumped ECCS Flowv Rate B 4.222-in Break D-12 TOODEE2 Clad Temperature B 4.222-in Break D-13 Core Mixture Level B 4.222-in Break D-14 Dowvncomer Liquid Level B 4.222-in Break D-15 Hot Assembly Steam Temperatures B 4.222-in Break D-16 Hot Assembly Heat Transfer Coefficients B 4.222-in Break D-17 Condensation Rate in Cold Leg Discharge B 4.222-in Break D-18 Break Quality B 4.222-in Break E-l Primary and Secondary System Pressures B 3.763-in Break E-2 Hot Assembly Region Void Fractions B 3.763-in Break E-3 Central Core Region Void Fractions B 3.763-in Break E-4 Average Core Region Void. Fractions B 3.763-in Break E-5 Upper Plenum Liquid Fraction B 3.763-in Break Attachment I to TXX-06125 Page 31 of 216 E-6 Hot Assembly.

Collapsed Water Level B 3.763 -in Break E-7 Hot Assembly Clad Temperatures B 3.763-in Break E-8 Loop Seal Void Fractions B 3.763-in Break E-9 Accumulator Mass Flow Rates B 3.763-in Break E-10 Break Flow Rate B 3.763-in Break E-1 I Total Pumped EGGS Flow Rate B 3.763-in Break E-12 TOODEE2 Clad Temperature B 3.763-in Break E-13 Core Mixture Level B 3.763-in Break E-14 Downcomer Liquid Level B 3.763-in Break E-15 Hot Assembly Steam Temperatures B 3.763-in Break E-16 Hot Assembly Heat Transfer Coefficients B 3.763-in Break E-17 Condensation Rate in Cold Leg Discharge B 3.763-in Break E-18 Break Quality B 3.763-in Break F-i Primary and Secondary System Pressures B 1.5-in Break (no BL SI)F-2 Hot Assembly Region Void Fractions B 1.5-in Break (no BL SI)F-3 Central Core Region Void Fractions B 1.5-in Break (no BL SI)F-4 Average Core Region Void Fractions B 1.5-in Break (no BL SI)F-5 Upper Plenum Liquid Fraction B 1.5-in Break (no BL SI)F-6 Hot Assembly Collapsed Water Level B 1.5-in Break (no BL SI)F-7 Hot Assembly Clad Temperatures B 1.5-in Break (no BL SI)F-8 Loop Seal Void Fractions B 1.5-in Break (no BL, SI)F-9 Accumulator Mass Flow Rates B 1.5-in Break (no BL SI)F-10 Break Flow Rate B 1.5-in Break (no BL, SI)F-IlI Total Pumped EGGS Flow Rate B 1.5-in Break (no BL SI)F- 12 TOODEE2 Clad Temperature B 1.5-in Break (no BL SI)F-13 Core Mixture Level B 1.5-in Break (no Bl, SI)F- 14 Downcomer Liquid Level B 1.5-in Break (no BL SI)F-IS Hot Assembly Steam Temperatures B 1.5-in Break (no BL SI)F-16 Hot Assembly Heat Transfer Coefficients B 1.5-in Break (no BL SI)F-17 Condensation Rate in Cold Leg Discharge B 1.5-in Break (no BL SI)F-18 Break Quality B 1.5-in Break (no BL SI)G-1 Primary and Secondary System Pressures B 6-in Break (no BL SI)G-2 Hot Assembly Region Void Fractions B 6-in Break (no BL SI)G-3 Central Core Region Void Fractions B 6-in Break (no BL SI)G-4 Average Core Region Void Fractions B 6-in Break (no BL SI)G-5 Upper Plenum Liquid Fraction B 6-in Break (no BL SI)G-6 Hot Assembly Collapsed Water Level B 6-in Break (no BL SI)G-7 Hot Assembly Clad Temperatures B 6-in Break (no BL SI)G-8 Loop Seal Void Fractions B 6-in Break (no BL SI)G-9 Accumulator Mass Flow Rates B 6-in Break (no BL SI)G-10 Break Flow Rate B 6-in Break (no BL, SI)G-l 1 Total Pumped EGGS Flow Rate B 6-in Break (no BL SI)G-12 TOODEE2 Clad Temperature B 6-in Break (no BL SI)

Attachment I to TXX-06125 Page 32 of 216 G-13 Core Mixture Level B 6-in Break (no BL, SI)G- 14 Downcomer Liquid Level B 6-in Break (no BL SI)G- 15 Hot Assembly Steam Temperatures B 6-in Break (no BL, SI)G-l16 Hot Assembly Heat Transfer Coefficients B 6-in Break (no BL, SI)G-17 Condensation Rate in Cold Leg Discharge B 6-in Break (no BL SI)G-18 Break Quality B 6-in Break (no BL, SI)H-i Primary and Secondary System Pressures B 4-in Break (Only broken LS cleared)H-2 Hot Assembly Region Void Fractions B 4-in Break (Only broken LS cleared)H-3 Central Core Region Void Fractions B 4-in Break (Only broken LS cleared)H-4 Average Core Region Void Fractions B 4-in Break (Only broken LS cleared)H-5 Upper Plenum Liquid Fraction B 4-in Break (Only broken LS cleared)H-6 Hot Assembly Collapsed Water Level B 4-in Break (Only broken LS cleared)H-7 Hot Assembly Clad Temperatures B 4-in Break (Only broken LS cleared)H-8 Loop Seal Void Fractions B 4-in Break (Only broken LS cleared)H-9 Accumulator Mass Flow Rates B 4-in Break (Only broken LS cleared)H-10 Break Flow Rate B 4-in Break (Only broken LS cleared)H- 11 Total Pumped ECCS Flow Rate B 4-in Break (Only broken LS cleared)H-12 TOODEE2 Clad Temperature B 4-in Break (Only broken LS cleared)H-13 Core Mixture Level B 4-in Break (Only broken LS cleared)H-14 Downcomer Liquid Level B 4-in Break (Only broken LS cleared)H-IS Hot Assembly Steam Temperatures B 4-in Break (Only broken LS cleared)H-16 Hot Assembly Heat Transfer Coefficients B 4-in Break (Only broken LS cleared)H-17 Condensation Rate in Cold Leg Discharge B 4-in Break (Only broken LS cleared)H-18 Break Quality B 4-in Break (Only broken LS cleared)I-1 Primary and Secondary System Pressures B 4-in Break (No HL Leakage Path)1-2 Hot Assembly Region Void Fractions B 4-in Break (No HL Leakage Path)1-3 Central Core Region Void Fractions B 4-in Break (No HL Leakage Path)1-4 Average Core Region Void Fractions B 4-in Break (No HL Leakage Path)1-S Upper Plenum Liquid Fraction B 4-in Break (No HL Leakage Path)1-6 Hot Assembly Collapsed Water Level B 4-in Break (No HL Leakage Path)1-7 Hot Assembly Clad Temperatures B 4-in Break (No HL Leakage Path)1-8 Loop Seal Void Fractions B 4-in Break (No HL Leakage Path)1-9 Accumulator Mass Flow Rates B 4-in Break (No HL Leakage Path)1-10 Break Flow Rate B 4-in Break (No HL Leakage Path)I-I I Total Pumped ECCS Flow Rate B 4-in Break (No HL Leakage Path)1-12 TOODEE2 Clad Temperature B 4-in Break (No HL Leakage Path)1-13 Core Mixture Level B 4-in Break (No HL Leakage Path)1-14 Downcomer Liquid Level B 4-in Break (No HL Leakage Path)1-IS Hot Assembly Steam Temperatures B 4-in Break (No HL Leakage Path)1-16 Hot Assembly Heat Transfer Coefficients B 4-in Break (No HL Leakage Path)1-17 Condensation Rate in Cold Leg Discharge B 4-in Break (No HL Leakage Path)1-18 Break Quality B 4-in Break (No HL Leakage Path)

Attachment 1 to TXX-06 125 Page 33 of 216 J-1 Primary and Secondary System Pressures B 4-in Break (Renodalized loop seals)J-2 Hot Assembly Region Void Fractions B 4-in Break (Renodalized loop seals)J-3 Central Core Region Void Fractions B 4-in Break (Renodalized loop seals)J-4 Average Core Region Void Fractions B 4-in Break (Renodalized loop seals)J-5 Upper Plenum Liquid Fraction B 4-in Break (Renodalized loop seals)J-6 Hot Assembly Collapsed Water Level B 4-in Break (Renodalized loop seals)J-7 Hot Assembly Clad Temperatures B 4-in Break (Renodalized loop seals)J-8 Loop Seal Void Fractions B 4-in Break (Renodalized loop seals)J-9 Accumulator Mass Flow Rates B 4-in Break (Renodalized loop seals)J-1 0 Break Flow Rate B 4-in Break (Renodalized loop seals)J-1 I Total Pumped ECCS Flow Rate B 4-in Break (Renodalized loop seals)J-12 TOODEE2 Clad Temperature B 4-in Break (Renodalized loop seals)J-13 Core Mixture Level B 4-in Break (Renodalized loop seals)J-14 Downcomer Liquid Level B 4-in Break (Renodalized loop seals)J-15 Hot Assembly Steam Temperatures B 4-in Break (Renodalized loop seals)J-16 Hot Assembly Heat Transfer Coefficients B 4-in Break (Renodalized loop seals)J-17 Condensation Rate in Cold Leg Discharge B 4-in Break (Renodalized loop seals)J-18 Break Quality B 4-in Break (Renodalized loop seals)

Attachment I to TXX-06 125 Page 34 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break 2500.0 -_ _ _ _ ____ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ____ ____--P 174010000*P 570010000 S2000.0- P57000--_0. x P 572010000 0 u P 573010000 S 1500 .0 -------- ----- -------- ---- --ILI La 0 1 0 0 0 .0-- -- -------------

--- ---------------

U)2-1 0. 50.0 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Time (sec)Figure A-I Primary and Secondary System Pressures in Break Attachment I to TXX-06125 Page 35 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break 1.2- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _-_ _ _ _-4o-VOIDG 122010000 aVOIDG 117010000 VOIDG 112010000 1.0 ----- ----------0.8 ---0~0.2 ----0 200 400 600 800 1000 1200 1400 1600 1800 2000 Time (sec)Figure A-2 Hot Assembly Region Void Fractions in Break Attachment I to TXX-06 125 Page 36 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break 1.2 1.0 0.8 0 UL0.6 0.4 0.2 0.0 0 200 400 600 800 1000 1200 1400 1600 Time (sec)Figure A-3 Central Core Region Void Fractions in Break 1800 2000 cc;;

Attachment I to TXX-06 125 Page 37 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break 1.2 1.0 0.8 0 0.4 0.6 0.0 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Time (sec)Figure A-4 Average Core Region Void Fractions in Break GC, 0ýr Attachment I to TXX-06 125 Page 38 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break 1.2 1.0.2 0.8 IL o 0.6 IL 0.4 0.2 0.0 0 200 400 600 800 1000 1200 1400 1600 1800 Time (sec)2000 Figure A-5 Upper Plenum Liquid Fraction in Break C-Q5D Attachment 1 to TXX-06 125 Page 39 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-nch Break 0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 Koa 4 +/ + 4- 4 F + +/- 4 4- 4 ______i F -% ----- 44 114 ---- -- L --I AJW 4 4 ~-4 4- -4 ~ 4- _______N If Opro,__ __ __ ___ I. __0 200 400 600 800 1000 Time (sec)1200 1400 1600 1800 2000 Figure A-6 Hot Assembly Collapsed Water Level in Break Attachment I to TXX-06 125 Page 40 of 216 GPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break 1300.0 1200.0 1100.0 L-9.- 1000.0 900.CL E 0)I-800.0 700.0 600.0 500.0 400.0 0 200 400 600 800 1000 1200 1400 1600 1800 Time (sec)2000 Figure A-7 Hot Assembly Clad Temperatures in Break coC9 Attachment 1 to TXX-06 125 Page 41 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break 1.2 1.0 1-itu.2 0.8 U.U-0 Co0.0.2 0.0--.--VOIDG 460030000* VOIDG 461030000 VOIDG 462030000* VOIDG 463030000 a 3 U U I,'x x x ix K K X X 0 200 400 600 800 1000 Time (sec)1200 1400 1600 1800 2000 Figure A-8 Loop Seal Void Fractions in Break Attachment 1 to TXX-06 125 Page 42 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break 50.0 40.0-sMFLOWJ 735000000

'* MFLOWJ 736000000 MFLOWJ 737000000* MFLOWJ 738000000 E.D30.0 0'U20.0 0 E 100o nxA--4 ft r 18~00 20 1~0.0-10.0-p --mm. ininm 11 h i 1200 400 600 800 1000 1400 14,00 16100 Time (sec)Figure A-9 Accumulator Mass Flow Rates in Break Attachment 1 to TXX-06 125 Page 43 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break 1200.0 1000.0 800.0 600.0 400.0 E 0 a: J9 I..200.0 0.0 0 200 400 600 800 1000 1200 1400 1600 1800 Time (sec)2000 Figure A-1 Break Flow Rate in Break Attachment 1 to TXX-06 125 Page 44 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break E 0 il: cn 0 100.0 90.0 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 0 200 400 600 800 1000 1200 1400 Time (sec)Figure A-IlI Total Pumped ECCS Flow Rate in Break 1600 1800 2000 Attachment 1 to TXX-06 125 Page 45 of 216 1300 1200 1100 L 900 S800 7 0 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break-PCT Node (10.125 ft)___No Rupture Predicted 600 500 400 0 200 400 600 800 1000 Time (sec)1200 1400 1600 1800 2000 Figure A-12 TOODEE2 Clad Temperature in Break Attachment 1 to TXX-06 125 Page 46 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break-J 0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 0 200 400 600 800 1000 1200 1400 1600 1800 Time (sec)2000 Figure A-13 Core Mixture Level in Break Attachment I to TXX-06 125 Page 47 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break 30.0 25.0> 20.0-J CL 0.E 5.0 0 5.0 0.0 0 200 400 600 800 1000 1200 1400 1600 1800 Time (sec)2000 Figure A-14 Downcomer Liquid Level in Break Attachment 1 to TXX-06 125 Page 48 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break 900.0 850.0 800.0 OLL 750.0*9700.0 E 650.0 E600.0 (U (550.0 500.0 450.0 400.0 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Time (sec)Figure A-15 Hot Assembly Steam Temperatures in Break cIF1 Attachmnent 1 to TXX-06 125 Page 49 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break 4.0 3.5 0, S3.0 C 2.5 a)o 2.0 4-S1.5 4)1.0 0) .0 0 200 400 600 800 1000 1200 1400 1600 1800 Time (sec)2000 Figure A-16 Hot Assembly Heat Transfer Coefficients in Break C 10 Attachment 1 to TXX-06 125 Page 50 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break 0.0 S-0.10-0 .2 --- -----------------

C -0.20490100

-.VAPGEN 490010000 0-0.4 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Time (sec)Figure A-Il7 Condensation Rate in Cold Leg Discharge in Break Attachment 1 to TXX-06 125 Page 51 of 216 CPSES-1 SBLOCA D76 RSG Analysis 3-Inch Break 1.0-0.9_ _ _ _ ___ _ __ _ _ _ _ _ _0.8-_ _ _ _ _ ___ __0.7-_ _ _ ___ _>%' 0.6.1.X c~0.4-___

_ __ _ __0.3__ _ _ _ __1_ _ _ _0.2 _ _ _ _ _ _ _ _ __1_ _ _ _0.1 _____0 200 400 600 800 1000 1200 1400 1600 1800 2000 Time (sec)Figure A-18 Break Quality in Break Attachment 1 to TXX-06 125 Page 52 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break 2500.0-.P 174010000*P 570010000 La00. P 571010000 0. P 572010000 aP 573010000 S 1500.0 ------- ---CL VU_ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _0 0.G1000 20 0 0 0 0010 Time (sec)Figure B-I Primary and Secondary System Pressures in Break Attachment I to TXX-06 125 Page 53 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break 1.2 1.0 0.8 0.2 0.0 0 200 400 600 800 1000 Time (sec)Figure B-2 Hot Assembly Region Void Fractions in Break 1200 Attachment I to TXX-06 125 Page 54 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break 1.2-.-VOIDG 140010000 uVOIDG 135010000 VOIDG 130010000 1.0 .~ .. .. ... ------- I 0.8-0U 0 .4 ---------

---

-~ -- ----- ---........

0.03 0. 200 400 600 800 1000 1200 Time (sec)Figure B-3 Central Core Region Void Fractions in Break Attachment 1 to TXX-06 125 Page 55 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break 1.2-.o-VOIDG 160010000 aVOIDG 155010000 VOIDG 150010000 1.0 -_ _ _ _ _ _ _ _ _ _ _ _0 .8 ------ ----0.U.90 0.0 -----------

--- ------ -- -----------------------------------------------

---

--__ _ __ _ _ _ -__ _ __ _ _0 200 400 600 800 1000 1200 Time (sec)Figure B-4 Average Core Region Void Fractions in Break Attachment 1 to TXX-06 125 Page 56 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break 1.2 1 .0 -- ----------------- -----.-VOIDF 166010000 O 0.8 *VOIDF 173010000 a. 0.4 --- ----------

---_0.0-0 20 40 6008001000120 Time (see)Figure B-5 Upper Plenum Liquid Fraction in Break Attachment I to TXX-06 125 Page 57 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break 14.0 12.0 10.0 0.0 8.0 6.0 4.0 2.0 0.0 0 200 400 600 800 1000 Time (sec)1200 Figure B-6 Hot Assembly Collapsed Water Level in Break Attachment 1 to TXX-06 125 Page 58 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break 2000.0 1800.0------------

-'HTTEMP 129101108*HTTEMP 129101208 1600.0 HTEM 121030 -------xHTTEMP 129101408 S1400.0 ------ -(D CL 1000 8 1 00.0 ---- ----_ --------1 0 000 ---~ ------ -----------

-- -400.0 0 200 400 600 800 1000 1200 Time (sec)Figure B-7 Hot Assembly Clad Temperatures in Break Attachment 1 to TXX-06 125 Page 59 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break 1.2 1.0 ---- ----Cx.2 0.8 ---VOD 6000*--VOIDG 460030000 0 .----. ....... --- --- -- -------> VOIDG 461030000 0.6 VOIDG 462030000 0 0 0 .4 -- ---- ---- -- ---- -- -- -- -- ------ ----0.2 ---------0.0. _______0 200 400 600 800 1000 1200 Time (sec)Figure B-8 Loop Seal Void Fractions in Break Attachment 1 to TXX-06 125 Page 60 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break 120.0 100.0 E.0 0 4.0 E 80.0 60.0 40.0 20.0-.-oMFLOWJ 735000000* MFLOWJ 736000000 MFLOWJ 737000000* MFLOWJ 738000000*A A A U: ~, A I U F,'4--~ ~-4 0.0-20.0 U ~ I ~ mm-I 12 00 200 400 600 800 1600o Time (sec)Figure B-9 Accumulator Mass Flow Rates in Break Attachment 1 to TXX-06 125 Page 61 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break 2000.0 1800.0 1600.0 1400.0 E 1200.0 01000.0 S800.0 600.600.0 400.0 0.0 0 200 400 600 800 1000 Time (sec)Figure B-10 Break Flow Rate in Break 1200 Attachment I to TXX-06 125 Page 62 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break E 0 0 100.0 90.0 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 0 200 400 600 800 Time (sec)Figure B-1lI Total Pumped ECCS Flow Rate in Break 1000 1200 Attachment 1 to TXX-06 125 Page 63 of 216 2000 1800 1600 L-2" 1400 CL E 1200 800 600 400 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break_______________ -PCT Node (1 0.125 ft) _____-No Rupture Predicted 0 200 400 600 Time (sec)800 1000 1200 Figure B-12 TOODEE2 Clad Temperature in Break Attachment I to TXX-06 125 Page 64 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break 14.0 12.0 10.0 0 0 UJ 8.0 6.0 4.0 2.0 0.0 0 200 400 600 800 1000 Time (sec)1200 Figure B-13 Core Mixture Level in Break Attachment I to TXX-06 125 Page 65 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break 30.0 25.0> 20.0-J 0.E 5.0 10.0 0 5.0 0.0_ I _ I _____ ____ I ____ f ____ I ____ ____0 200 400 600 Time (sec)800 1000 1200 Figure B-14 Downcomer Liquid Level in Break Attachment 1 to TXX-06 125 Page 66 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break 1600.0 1400.0 CL E 0 E MU 1200.0 1000.0 800.0 600.0 400.0 0 200 400 600 800 1000 Time (sec)Figure B-15 Hot Assembly Steam Temperatures in Break 1200 C-2r, Attachment 1 to TXX-06 125 Page 67 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break-.+-HTHTC 1291011 uHTHTC 1291012 U.Q 4 HTHTC 1291014~2.5 -- ----- --- -----C.4"'01.0.0 -- 1p 0 200 400 600 800 1000 1200 Time (sec)Figure B-16 Hot Assembly Heat Transfer Coefficients in Break Attachment 1 to TXX-06 125 Page 68 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break 0.00 CI, r--0.10 E.0.2 0)i-0.20 0 0-0.60 0 0200 400 600 800 1000 Time (sec)Figure B-I 7 Condensation Rate in Cold Leg Discharge in Break 1200 Attachment 1 to TXX-06125 Page 69 of 216 CPSES-1 SBLOCA D76 RSG Analysis 4-Inch Break 1.0-0.91_ _- -_ --__0.8-1_ _ _ _M 0.4-___ ___ _______0.2 _ _ _ _ _ _ 1 0.1~0.0I 0 200 400 600 800 1000 1200 Time (sec)Figure B-18 Break Quality in Break Attachment 1 to TXX-06 125 Page 70 of 216 CPSES-1 SBLOCA D76 RSG Analysis 5-Inch Break 2500.0 M' 2000.0 16.(A 2 1500.0 CU 0 S1000.0 ad E I. 500.0 0.0 0 200 400 600 800 1000 Time (sec)Figure C-I Primary and Secondary System Pressures in Break 1200 Attachment I to TXX-06 125 Page 71 of 216 CPSES-1 SBLOCA D76 RSG Analysis 5-Inch Break 1.2--VOIDG 122010000*VOIDG 117010000 VOIDG 112010000 1 .0 -- --- --- --- ------- --- -- --~ -- ----0.8 -WI 0.2 0.64 0 20 40 6008001000120 Time (sec)Figure C-2 Hot Assembly Region Void Fractions in Break Attachment 1 to TXX-06 125 Page 72 of 216 CPSES-1 SBLOCA D76 RSG Analysis 5-Inch Break 1.2 _ _-4$-VOIOG 140010000*VOIOG 135010000 VOIDG 130010000 1.0 ---- ---- -- --- -- -----0.8 --0.6 ----M 0.4 ------0.2 0.0- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _0 200 400 600 800 1000 1200 Time (sec)Figure C-3 Central Core Region Void Fractions in Break Attachment I to TXX-06 125 Page 73 of 216 CPSES-1 SBLOCA D76 RSG Analysis 5-Inch Break 1.2 1.0 0.8 0 0.4 0.6 0.0 0 200 400 600 800 1000 Time (sec)1200 Figure C-4 Average Core Region Void Fractions in Break c24-Attachment I to TXX-06 125 Page 74 of 216 CPSES-1 SBLOCA D76 RSG Analysis 5-Inch Break 1.2 1.0 02 0.8 00.6 IL 0.4 0.2 0.0 0 200 400 600 800 1000 1200 Time (sec)Figure C-5 Upper Plenum Liquid Fraction in Break c2-7